* Use distances on all valid paths rather than just shortest distance.
* Optimise BondPaths.
* Optimise BondPaths.
* Hash coded for the bond paths.
* Faster find all paths.
* Build in gcc working.
* Comment.
* Remove debugging code.
* Update GettingStartedInPython.rst.
* Now need to split the clique and keep the largest fragment.
Lots of warnings about how slow this is.
Split out long tests.
* Back out a lot of changes. Remove the distance check with singleLargestFrag when building modular product.
* Tidy code.
Update docstrings.
Add explanation to GettingStartedInPython.rst.
* Fix single fragment test.
* Response to review.
---------
Co-authored-by: David Cosgrove <david@cozchemix.co.uk>
* Attempt at completeRingsOnly in mces
* changed option name and added a test
* code review
* check aromaticity and bond type matches before checking ring equivalencies
* update if statement
* typo
* Initial results sort the wrong way round on number of atoms and bonds.
* Improve sort for singleLargestFrag.
* Response to review - don't sort the fragment sets twice when opts.singleLargestFrag is true.
---------
Co-authored-by: David Cosgrove <david@cozchemix.co.uk>
* Add option to specify minimum clique size directly.
* Make the minCliqueSize uint and default to 0.
* Brain-fade.
* Send timeout and 'too many bonds' message to BOOST_LOG rather than std::cout, std::cerr respectively.
---------
Co-authored-by: David Cosgrove <david@cozchemix.co.uk>
* Fix atom ordering in bond label.
* Response to review.
---------
Co-authored-by: David Cosgrove <david@cozchemix.co.uk>
Co-authored-by: Greg Landrum <greg.landrum@gmail.com>
* Allow fragments of aromatic rings to match.
* Use existing code for checking bond is in ring.
---------
Co-authored-by: David Cosgrove <david@cozchemix.co.uk>
* New option - exactAtomTypeMatch.
* Python wrapper.
* REQUIRE(res.size()) not CHECK. If it fails, you get seg fault on subsequent tests otherwise.
* Change option name.
* C++ implementation of equivalent atoms via SMARTS.
* Python wrapper.
* Tidy.
* Check for too many classes.
* Better handling of spaces in split.
* Pass string by reference. Doh.
* Add ignoreBondOrders option.
* Fix bug where different atom orders gave different MCES sizes.
* Update function description.
---------
Co-authored-by: David Cosgrove <david@cozchemix.co.uk>
* enable Werror on Mac and Linux
* do not fail on boost multiprecision pessimizing move
* fix eigen array_bounds warning
* Fix unused arg in Rascal MCS
* fix range-loop-construct warning in Rascal MCES
* fix sign mismatched comparison
* drop unused lambda capture
* allow FMCS timeout test more time under Debug (not a warning!)
* fix fwd declaration of struct RascalClusterOptions
* fix deallocator mismatch
* fix two minor leaks
* fix a real leak
* more minor leaks
* fix another real leak, plus some potential ones
* fix std::move preventing copy ellision
* allow longer run time for debug builds
* make maxBondMatchPairs and getLargestFragSize unsigned int
* make snake case camel case
* update to current master, fix new warnings
* update again and more fixes
* add #include <optional>
* fix char array deallocation
* update and fixes in Marvin writer
* unsigned int
* more copy ellision fixes
* more copy ellision fixes, and typos
* and some more
